{
  "id": "rlvr",
  "code": "PTL-0156",
  "term": "Reinforcement Learning with Verifiable Rewards",
  "aliases": [
    "RLVR",
    "RL from verifiable rewards",
    "outcome-based reinforcement learning"
  ],
  "category": "test-time",
  "definition": "Reinforcement learning with verifiable rewards (RLVR) trains a language model with rewards computed by an automatic check of correctness, such as whether a math answer matches the reference or code passes its tests, instead of by a learned reward model.",
  "description": "Verifiable rewards are hard to game in the ways a learned reward model can be, and large-scale RL with them is how reasoning models learn to produce long, self-correcting chains of thought. The approach is limited to tasks with checkable answers, so open-ended writing and judgment still rely on preference-based rewards.",
  "example": null,
  "broader": [],
  "narrower": [],
  "related": [
    "reasoning-model",
    "rlhf",
    "reward-model",
    "reward-hacking",
    "self-taught-reasoner"
  ],
  "introduced": 2024,
  "added": "2026-10-10",
  "sources": [
    {
      "title": "Tulu 3: Pushing Frontiers in Open Language Model Post-Training",
      "authors": "Lambert et al.",
      "year": 2024,
      "url": "https://arxiv.org/abs/2411.15124"
    },
    {
      "title": "DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning",
      "authors": "DeepSeek-AI et al.",
      "year": 2025,
      "url": "https://arxiv.org/abs/2501.12948"
    }
  ],
  "url": "https://protologue.com/t/rlvr/",
  "citation": "Protologue. (2026). Reinforcement Learning with Verifiable Rewards. In Protologue: A Taxonomy of Prompting and LLM Techniques (v1.4.0, PTL-0156). https://protologue.com/t/rlvr/"
}